{"title":"AI Text Generators and Text Producers","authors":"Henrik Køhler Simonsen","doi":"10.1109/ICALT55010.2022.00071","DOIUrl":null,"url":null,"abstract":"AI-generated text production is becoming increasingly important in many industries, and it has already brought about dramatic changes in the ways we write texts and generate content. The article draws on empirical data from a descriptive-analytical study involving 70 test subjects. The population comprised 115 test persons, who received an e-mail with instructions. A sample of 70 test subjects participated in the study. First, the test subjects were asked to test a specific AI text generator (ATG) and conduct three prompting operations with the same linguistic content. Second, having tested the ATG, the test subjects were asked to participate in a questionnaire with ten questions focusing on how they experienced the performance of the ATG and how they worked with the ATG. The majority of the test subjects found that the tested ATG was easy to use when producing texts. When asked about the perceived quality of the AI-generated content, the respondents were not impressed with the quality and indicated that they needed to perform several editing operations. The data also indicate that ATGs need help before, during and after. This paper presents a three-phase editing framework, which can be used when using and teaching ATGs.","PeriodicalId":221464,"journal":{"name":"2022 International Conference on Advanced Learning Technologies (ICALT)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2022 International Conference on Advanced Learning Technologies (ICALT)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICALT55010.2022.00071","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1
Abstract
AI-generated text production is becoming increasingly important in many industries, and it has already brought about dramatic changes in the ways we write texts and generate content. The article draws on empirical data from a descriptive-analytical study involving 70 test subjects. The population comprised 115 test persons, who received an e-mail with instructions. A sample of 70 test subjects participated in the study. First, the test subjects were asked to test a specific AI text generator (ATG) and conduct three prompting operations with the same linguistic content. Second, having tested the ATG, the test subjects were asked to participate in a questionnaire with ten questions focusing on how they experienced the performance of the ATG and how they worked with the ATG. The majority of the test subjects found that the tested ATG was easy to use when producing texts. When asked about the perceived quality of the AI-generated content, the respondents were not impressed with the quality and indicated that they needed to perform several editing operations. The data also indicate that ATGs need help before, during and after. This paper presents a three-phase editing framework, which can be used when using and teaching ATGs.